Routing in ad-hoc sensor
networks has been usually addressed as a discrete optimization problem with
optimality criteria related to node energy minimization, time-delay through
the network, bandwidth maximization [18, 52, 54], etc. Congestion control is
a relatively new research topic for sensor networks, but has been investigated
extensively in the context of static (wired) network protocols. The congestion
control problem consists of setting both the source and link controls of the
network to ensure that the data rates and node bu?®ers remain stable and
stabilize to their optimal values. Various feedback schemes such as the primal
(dynamic source rates, static link rates), dual (static source rates, dynamic
link rates), and primal-dual (dynamic source rates, dynamic link rates) algorithms
have been proposed [35]. A general passivity framework is proposed in
[55] as a superset solution for flow control.
36
Chapter 2 Algorithms for Robotic Deployment of WSN
Several wireless network protocols have been proposed and implemented
using many of the above concepts, such as IEEE 802.11a,b, Bluetooth and
PAMAS, and some of them specifically apply to sensor networks, such as PICONET
and SMAC [11, 52, 54]. There are significant di?®erences between wireless
protocols for distributed sensors, and wireless protocols used for telecom
or computer networking. For example, concepts such as bandwidth fairness
between users do not apply in the sensor network case.
Pages:
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98